--- title: "SkyPilot Job Groups: Run RL on Heterogenous Hardware" url: https://daily.dev/posts/skypilot-job-groups-run-rl-on-heterogenous-hardware-komwla8ap source_url: https://blog.skypilot.co/job-groups/ type: article source: "SkyPilot" published: 2026-03-03T17:06:19.513Z updated: 2026-03-03T17:06:51.704Z tags: ["kubernetes", "distributed-systems", "reinforcement-learning"] reading_time: 7 upvotes: 0 comments: 0 language: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # SkyPilot Job Groups: Run RL on Heterogenous Hardware **[SkyPilot](https://daily.dev/sources/skypilot)** · 7 min read · 0 upvotes · 0 comments ## Summary SkyPilot Job Groups is a new feature that lets you define heterogeneous ML workloads in a single YAML file. It solves the coordination problem in RL post-training (GRPO, PPO, RLHF) where different components need different hardware: H100s for the policy trainer, cheaper GPUs for inference rollouts, and high-memory CPUs for replay buffers. Key capabilities include automatic provisioning of mixed instance types, DNS-based service discovery between tasks, declarative lifecycle management via primary/auxiliary task designation, and automatic recovery from preemptions. The same YAML runs on AWS, GCP, Azure, or Kubernetes without per-cloud configuration. Limitations include single-region constraint, Kubernetes requirement for DNS-based discovery, and fixed replica counts at submission time. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: --- Tags: [#kubernetes](https://daily.dev/tags/kubernetes), [#distributed-systems](https://daily.dev/tags/distributed-systems), [#reinforcement-learning](https://daily.dev/tags/reinforcement-learning) [View this post on daily.dev](https://daily.dev/posts/skypilot-job-groups-run-rl-on-heterogenous-hardware-komwla8ap)